Filtering

Join Our Community Subscribe to Paul's Posts

We’ve worked our way through the internal DAC chain to a point where we have finally decoded the numeric musical data into current steps, then converted those steps into voltage.

It is at this point that we wind up with our music back into the original form it started with but with one exception: there’s lots of little steps that have to be smoothed out and the remaining transitions removed. That’s the job of the analog filter.

Let’s start by taking a look at what we have to work with.

Note the top trace and the little stair steps that make up this sine wave. Once we run it through what is known as a low pass filter, we wind up with all the little steps removed and the result is a perfect looking sine wave as you can see in the lower trace.

How you make this filter can have a not so insignificant impact on the sonic performance of our DAC. In many cases, designers run the output of the current to voltage converter through yet another op amp forming what is known as an active filter. Yet another technique is to wrap the filter around the analog output stage and, yet a third method, is a passive filter. I prefer the passive filter approach.

Passive filters use only a few components outside the analog stage: capacitors and resistors typically. We found this approach to be sonically preferable when we starting building RIAA phono stages in our very first products. There we placed an entirely passive low pass filter between two active analog stages and achieved stunning results. In a DAC setting we do the same thing: placing a purely passive low pass filter between the current to voltage converter and the output analog stage.

This, in turn, allow us to build a beautifully clean and simple analog stage, the final leg of our DAC. That’s what we’ll focus on tomorrow.